Maximum Betweenness Centrality Research Articles

Resilience quantification and recovery strategy simulation for urban underground logistics systems under node and link attacks: A case study of Nanjing city

Urban Underground Logistics Systems (UULS) have become an emerging solution to mitigate urban surface traffic congestion, environmental pollution, and surface transport safety risks. However, during the operation of UULS, the use of advanced technologies such as the Internet of Things (IoT) introduces cybersecurity risks to the system. Moreover, severe natural disasters can also cause damage to underground transportation network links. Existing research and planning primarily concentrate on the system design and benefits of UULS, neglecting the system's service level under attack scenarios. This study outlines three representative UULS network prototypes and proposes a resilience quantification method centered on logistics efficiency. It also focuses on comparing the effectiveness of three recovery strategies. These strategies give priority to maximum flow, betweenness centrality, and regional importance, as well as the priority of node and link repairs. The resilience quantification method and recovery strategies are applied in a case study set in Nanjing City. The case study results reveal that the Two-echelon network shows exceptional resilience. Regarding the effectiveness of recovery strategies, the strategy based on maximum flow proves to be the most effective, and focusing on node repair can lead to higher system resilience. Based on these findings, this study offers recommendations to transportation and logistics management decision-makers, focusing on UULS resilience and recovery strategy selection. These recommendations are intended to provide valuable guidance for the planning and design of future UULS, ensuring their resilience and reliability.

Minimum spanning tree analysis of EEG resting-state functional networks in schizophrenia

Schizophrenia is a serious and complex mental disease, known to be associated with various subtle structural and functional deviations in the brain. Recently, increased attention is given to the analysis of brain-wide, global mechanisms, strongly altering the communication of long-distance brain areas in schizophrenia. Data of 32 patients with schizophrenia and 28 matched healthy control subjects were analyzed. Two minutes long 64-channel EEG recordings were registered during resting, eyes closed condition. Average connectivity strength was estimated with Weighted Phase Lag Index (wPLI) in lower frequencies: delta and theta, and Amplitude Envelope Correlation with leakage correction (AEC-c) in higher frequencies: alpha, beta, lower gamma and higher gamma. To analyze functional network topology Minimum Spanning Tree (MST) algorithms were applied. Results show that patients have weaker functional connectivity in delta and alpha frequency bands. Concerning network differences, the result of lower diameter, higher leaf number, and also higher maximum degree and maximum betweenness centrality in patients suggest a star-like, and more random network topology in patients with schizophrenia. Our findings are in accordance with some previous findings based on resting-state EEG (and fMRI) data, suggesting that MST network structure in schizophrenia is biased towards a less optimal, more centralized organization.

A deep learning model predicts the presence of diverse cancer types using circulating tumor cells

Circulating tumor cells (CTCs) are cancer cells that detach from the primary tumor and intravasate into the bloodstream. Thus, non-invasive liquid biopsies are being used to analyze CTC-expressed genes to identify potential cancer biomarkers. In this regard, several studies have used gene expression changes in blood to predict the presence of CTC and, consequently, cancer. However, the CTC mRNA data has not been used to develop a generic approach that indicates the presence of multiple cancer types. In this study, we developed such a generic approach. Briefly, we designed two computational workflows, one using the raw mRNA data and deep learning (DL) and the other exploiting five hub gene ranking algorithms (Degree, Maximum Neighborhood Component, Betweenness Centrality, Closeness Centrality, and Stress Centrality) with machine learning (ML). Both workflows aim to determine the top genes that best distinguish cancer types based on the CTC mRNA data. We demonstrate that our automated, robust DL framework (DNNraw) more accurately indicates the presence of multiple cancer types using the CTC gene expression data than multiple ML approaches. The DL approach achieved average precision of 0.9652, recall of 0.9640, f1-score of 0.9638 and overall accuracy of 0.9640. Furthermore, since we designed multiple approaches, we also provide a bioinformatics analysis of the gene commonly identified as top-ranked by the different methods. To our knowledge, this is the first study wherein a generic approach has been developed to predict the presence of multiple cancer types using raw CTC mRNA data, as opposed to other models that require a feature selection step.

Finding groups with maximum betweenness centrality via integer programming with random path sampling

Finding groups with maximum betweenness centrality via integer programming with random path sampling

Graph Analysis of Functional Brain Topology Using Minimum Spanning Tree in Subjective Cognitive Decline.

Subjects with subjective cognitive decline (SCD) are proposed as a potential population to screen for Alzheimer's disease (AD). Investigating brain topologies would help to mine the neuromechanisms of SCD and provide new insights into the pathogenesis of AD. Objectively cognitively unimpaired subjects from communities who underwent resting-state BOLD-fMRI and clinical assessments were included. The subjects were categorized into SCD and normal control (NC) groups according to whether they exhibited self-perceived cognitive decline and were worried about it. The minimum spanning tree (MST) of the functional brain network was calculated for each subject, based on which the efficiency and centrality of the brain network organization were explored. Hippocampal/parahippocampal volumes were also detected to reveal whether the early neurodegeneration of AD could be seen in SCD. A total of 49 subjects in NC and 95 subjects in SCD group were included in this study. We found the efficiency and centrality of brain network organization, as well as the hippocampal/parahippocampal volume were preserved in SCD. Besides, SCD exhibited normal cognitions, including memory, language, and execution, but increased depressive and anxious levels. Interestingly, language and execution, instead of memory, showed a significant positive correlation with the maximum betweenness centrality of the functional brain organization and hippocampal/parahippocampal volume. Neither depressive nor anxious scales exhibited correlations with the brain functional topologies or hippocampal/parahippocampal volume. SCD exhibited preserved efficiency and centrality of brain organization. In clinical practice, language and execution as well as depression and anxiety should be paid attention in SCD.

An effective edge-adding strategy for enhancing network traffic capacity

With the rapid development of technology, the number of packets that need to be transmitted in the network is increasing, which leads to network congestion and reduces the user experience. Therefore, it is necessary to adjust the transmission path of packets in the network to improve the transmission threshold of packets in the network. In this paper, we abstract the real world network into a scale-free network for research, using nodes to represent individuals in reality, and edges to represent connections between individuals. Scale-free networks have serious inhomogeneity. Nodes in scale-free networks cannot be connected to all nodes in the network, which makes some gaps in the network. “Structural hole” describes the gaps in the network, which make some nodes have certain controllability to the whole network. But it only considers the relationship between nodes and their neighbors, without considering the overall structure of the whole network topology. K-shell decomposition algorithm can efficiently and accurately identify the location of nodes in the network. Therefore, K-shell decomposition algorithm is a global index. However, it still has shortcomings that does not consider the topological relationship between neighbors. Both structural hole and K-shell theory can characterize the importance of nodes. The combination of the two can make up for each other’s flaws and make the importance of nodes in the network more even. The more even the node importance, the more even the load of the node, the greater the increase in network traffic capacity. In this paper, we propose a network edge-adding strategy combining the structural hole theory and the k-shell algorithm to improve the traffic capacity. Extensive simulations have been performed to estimate the effectiveness of the proposed method under the efficient routing strategy. According to the simulation results, we can see that when the network size is fixed, regardless of the average degree of nodes, our proposed strategy improves the network traffic capacity, reduces the maximum betweenness centrality of nodes, and makes the load of nodes in the network more average. At the same time, our strategy improves the robustness of the network.

Network-Based Differences in the Vaginal and Bladder Microbial Communities Between Women With and Without Urgency Urinary Incontinence.

BackgroundLittle is known about the relationship of proximal urogenital microbiomes in the bladder and the vagina and how this contributes to bladder health. In this study, we use a microbial ecology and network framework to understand the dynamics of interactions/co-occurrences of bacteria in the bladder and vagina in women with and without urgency urinary incontinence (UUI).MethodsWe collected vaginal swabs and catheterized urine specimens from 20 women with UUI (cases) and 30 women without UUI (controls). We sequenced the V4 region of the bacterial 16S rRNA gene and evaluated using alpha and beta diversity metrics. We used microbial network analysis to detect interactions in the microbiome and the betweenness centrality measure to identify central bacteria in the microbial network. Bacteria exhibiting maximum betweenness centrality are considered central to the microbe-wide networks and likely maintain the overall microbial network structure.ResultsThere were no significant differences in the vaginal or bladder microbiomes between cases and controls using alpha and beta diversity. Silhouette metric analysis identified two distinct microbiome clusters in both the bladder and vagina. One cluster was dominated by Lactobacillus genus while the other was more diverse. Network-based analyses demonstrated that vaginal and bladder microbial networks were different between cases and controls. In the vagina, there were similar numbers of genera and subgroup clusters in each network for cases and controls. However, cases tend to have more unique bacterial co-occurrences. While Bacteroides and Lactobacillus were the central bacteria with the highest betweenness centrality in controls, Aerococcus had the highest centrality in cases and correlated with bacteria commonly associated with bacterial vaginosis. In the bladder, cases have less than half as many network clusters compared to controls. Lactobacillus was the central bacteria in both groups but associated with several known uropathogens in cases. The number of shared bacterial genera between the bladder and the vagina differed between cases and controls, with cases having larger overlap (43%) compared to controls (29%).ConclusionOur study shows overlaps in microbial communities of bladder and vagina, with higher overlap in cases. We also identified differences in the bacteria that are central to the overall community structure.

Bacterial community succession and influencing factors for Imperata cylindrica litter decomposition in a copper tailings area of China

Litter decomposition is a critical component of the ecological nutritional transformation process. In a copper mining area, the litter from Imperata cylindrica is the major indicator for restoring heavy metal-polluted copper mining lands. Large amounts of litter are generated at the end of the plant growing season during the process of vegetation restoration in copper mining areas, and the microbial dynamics play an important role in soil nutrient turnover during the decomposition of litter. Investigating the characteristics and interactions of bacterial communities during litter decomposition will clarify the driving mechanisms of organic matter and nutrient cycling in copper mining areas that harbor contaminated soils. Here, we report the results of an in situ decomposition experiment that lasted for a total of 460 days from three of the 16 copper mining subdams with heavy metal pollution and different phytoremediation histories (e.g., 50, 22 and 5 years) to explore the bacterial communities as the driving factors of litter decomposition. The total carbon contents of the litter decreased by 62.6% and 71.5% in the decomposition process at those sites with phytoremediation histories of 50 and 22 years (S516 and S536), respectively, but decreased by only 25.8% at the site with a phytoremediation history of 5 years (S560). The optimal C/N ratios in the three different restoration stages varied and were 65.5, 86.7 and 39.3 in S516, S536, S560, respectively. Litter decomposition enriched the heavy metal contents such as cadmium, copper (Cu), lead and zinc (P < 0.05) in litter. Proteobacteria and Actinobacteriota were the dominant bacterial phyla during the different litter decomposition stages, which accounted for 91.66% of the relative abundances in the bacterial communities. Moreover, the role of Friedmanniella, which had the highest betweenness centrality (BC) value, was critical in sustaining both the structure and function of the bacterial communities during the early decomposition stage. However, Quadrisphaera, with the maximum BC value (1074.8), became the dominant genus as litter decomposition progressed. The most crucial factors that affected the litter bacterial communities were the litter pH and copper contents. The obtained results will be helpful to provide a further understanding of litter decomposition mechanisms and will provide a scientific basis for improving the effectiveness of material circulation and nutrient transformation in degraded copper mining ecosystems.

EEG Alpha and Beta Band Functional Connectivity and Network Structure Mark Hub Overload in Mild Cognitive Impairment During Memory Maintenance.

Background: While decreased alpha and beta-band functional connectivity (FC) and changes in network topology have been reported in Alzheimer’s disease, it is not yet entirely known whether these differences can mark cognitive decline in the early stages of the disease. Our study aimed to analyze electroencephalography (EEG) FC and network differences in the alpha and beta frequency band during visuospatial memory maintenance between Mild Cognitive Impairment (MCI) patients and healthy elderly with subjective memory complaints.Methods: Functional connectivity and network structure of 17 MCI patients and 20 control participants were studied with 128-channel EEG during a visuospatial memory task with varying memory load. FC between EEG channels was measured by amplitude envelope correlation with leakage correction (AEC-c), while network analysis was performed by applying the Minimum Spanning Tree (MST) approach, which reconstructs the critical backbone of the original network.Results: Memory load (increasing number of to-be-learned items) enhanced the mean AEC-c in the control group in both frequency bands. In contrast to that, after an initial increase, the MCI group showed significantly (p < 0.05) diminished FC in the alpha band in the highest memory load condition, while in the beta band this modulation was absent. Moreover, mean alpha and beta AEC-c correlated significantly with the size of medial temporal lobe structures in the entire sample. The network analysis revealed increased maximum degree, betweenness centrality, and degree divergence, and decreased diameter and eccentricity in the MCI group compared to the control group in both frequency bands independently of the memory load. This suggests a rerouted network in the MCI group with a more centralized topology and a more unequal traffic load distribution.Conclusion: Alpha- and beta-band FC measured by AEC-c correlates with cognitive load-related modulation, with subtle medial temporal lobe atrophy, and with the disruption of hippocampal fiber integrity in the earliest stages of cognitive decline. The more integrated network topology of the MCI group is in line with the “hub overload and failure” framework and might be part of a compensatory mechanism or a consequence of neural disinhibition.

The structure and behaviour of hierarchical infrastructure networks

Critical services depend on infrastructure networks for their operation and any disruption to these networks can have significant impacts on society, the economy, and quality of life. Such networks can be characterised as graphs which can be used to understand their structural properties, and the effect on their behaviour and robustness to hazards. Using a suite of graphs and critical infrastructure networks, this study aims to show that networks which exhibit a hierarchical structure are more likely to be less robust comparatively to non-hierarchical networks when exposed to failures, including those which supply critical services. This study investigates the properties of a hierarchical structure through identifying a set of key characteristics from an ensemble of graph models which are then used in a comparative analysis against a suite of spatial critical infrastructure networks. A failure model is implemented and applied to understand the implications of hierarchical structures in real world networks for their robustness to perturbations. The study concludes that a set of three graph metrics, cycle basis, maximum betweenness centrality and assortativity coefficient, can be used to identify the extent of a hierarchy in graphs, where a lack of robustness is linked to the hierarchical structure, a feature exhibited in both graph models and infrastructure networks.

Structural Heterogeneity and Evolutionary Dynamics on Complex Networks

The study of evolutionary games on networks has revealed the impact of population structure on evolutionary dynamics. Unlike the case in well-mixed population where defection is favored by natural selection, certain types of networks have shown to favor cooperation. However, most previous research work has been focusing on frequency-based analysis, and emphasized on the update strategy adopted by each player, and thus generally considered the group of players with the same strategy as a whole. While it is powerful in deriving analytic results using this approach, the heterogeneity of players within such groups is effectively overlooked. In this paper, we attempt to emphasize more on the heterogeneity of players that comes from the network structure in evolutionary dynamics. Particularly, the prestige of a player is represented by its centrality, and it is reflected in an adapted payoff function. We provide several viable centrality measures that can be calculated using the adjacency matrix of the network. The relation between different centrality measures of the invader and the fixation of cooperation is analyzed via computational simulations. Results show that in the proposed model, compared to other three centrality measures, invaders with maximum betweenness centrality have significant advantage in terms of the fixation probability of cooperation, in both scale-free and small-world networks.

On achieving capacity-enhanced small-world networks

Networks, whether it is communication or transportation, often suffer from significant degradation in average network flow capacity (ANFC) due to the presence of one or more bottleneck nodes. Presence of a few long-ranged links, connecting distant nodes in a regular network, enhances ANFC and incorporates small-world characteristics. However, the existing deterministic long-ranged link addition strategies based on minimum average path length, maximum betweenness centrality, or maximum closeness centrality, cannot guarantee significant improvement in ANFC. In this paper, we propose an exhaustive search-based long-ranged link (LL) addition algorithm, maximum flow capacity (MaxCap), which deterministically maximizes ANFC, based on the maximum flow (of information or objects) among node-pairs in the context of weighted undirected networks. Furthermore, based on the observations from MaxCap, we propose a new LL addition heuristic, average network flow capacity enhancement using small-world characteristics (ACES), which significantly enhances ANFC and the LL length-type product, and improves traffic load distribution in a weighted undirected network. We validate the performance of our LL addition method through exhaustive simulations on various arbitrary networks and real-world road transportation networks. ACES can be applied to many real-world applications in communication networks, transportation networks, and tactical networks where ANFC is a critical parameter.

Betweenness Centrality Based Software Defined Routing

Software-defined networking (SDN) enables routing control to program in the logically centralized controllers. It is expected to improve the routing efficiency even in highly dynamic situations. In this article, we make an in-depth observation of practical Internet datasets and investigate the relationship between betweenness centrality and network throughput . Furthermore, we propose a new routing observation factor, differential ratio of betweenness centrality (DRBC), to denote the varying amplitude of betweenness centrality to node degree. We reveal an interesting phenomenon that DRBC is proportional to the routing efficiency when the maximum betweenness centrality varies in a small range. Based on this, a DRBC-based routing scheme is proposed to improve routing efficiency. The experimental results verify that DRBC-based routing can improve the network throughput and accelerate the routing optimization.

TÜRKİYE’DE YEREL VE ULUSLARARASI HAVA TAŞIMACILIĞI YAPISININ AĞ ANALİZİ İLE İNCELENMESİ

Considering the effectiveness of air transportation, the investigation of the dynamics of flight routes are having a great attention, lately. Even several studies are conducted in a global scale, some hubs are required more consideration in inter-city and inter-country transportation. According to the study of [Song and Yeo], İstanbul located in Turkey has the maximum degree centrality among 1.060 airports and maximum betweenness centrality among the top 30 airports. In this study, we aim to see the air transportation network structure deeply for this country depending on its topographic characteristics. For this purpose, social network analysis is used as an analysis method and visualization tools to describe the correlation structure of the data. Thus, we illustrate some of the network-level and node-level metrics in aforementioned analysis by exploring route connections among airports. Additionally, the network maps are depicted to better understand the air routes structure. The results indicate that İstanbul has a huge impact among airports in terms of both domestic and international transportation depending on the empirical data of Turkey for the year of 2014. Classifying by countries, Germany and Cyprus has the largest connection measurement results with Turkey.

Abnormal Functional Brain Networks in Mild Cognitive Impairment and Alzheimer's Disease: A Minimum Spanning Tree Analysis.

Alzheimer's disease (AD) disrupts the topological architecture of whole-brain connectivity. Minimum spanning tree (MST), which captures the most important connections in a network, has been considered an unbiased method for brain network analysis. However, the alterations in the MST of functional brain networks during the progression of AD remain unclear. Here, we performed an MST analysis to examine the alterations in functional networks among normal controls (NCs), mild cognitive impairment (MCI) patients, and AD patients. We identified substantial differences in the connections among the three groups. The maximum betweenness centrality, leaf number, and tree hierarchy of the MSTs showed significant group differences, indicating a more star-like topology in the MCI patients and a more line-like topology in the NCs and AD patients. These findings may correspond to changes in the core of the functional brain networks. For nodal properties (degree and betweenness centrality), we determined that brain regions around the cingulate gyrus, occipital lobes, subcortex, and inferior temporal gyrus showed significant differences among the three groups and contributed to the global topological alterations. The leaf number and tree hierarchy, as well as the nodal properties, were significantly correlated with clinical features in the MCI and AD patients, which demonstrated that more star-to-line topology changes were associated with worse cognitive performance in these patients. These findings indicated that MST properties could capture slight alterations in network topology, particularly for the differences between NCs and MCI patients, and may be applicable as neuroimaging markers of the early stage of AD.

Finding clique clusters with the highest betweenness centrality

This article introduces and studies the most betweenness-central clique problem, which involves finding a clique cluster of maximum betweenness centrality in a connected network. This problem allows one to extend the traditional notion of finding a single most influential (or central) member, to one of identifying central groups whose members collectively satisfy a structural property that is meaningful relative to the underlying system. Among others, betweenness-central cliques find application in corporate, social, communication, power grid, and biological network analysis. In this study, the betweenness centrality measure adopted to define our problem is based on the summation of the percentages representing the ratio of the shortest paths between every pair of vertices in a network that intersect a given clique. The decision version of this problem is proven to be NP-complete, and it is shown that an optimal solution to this problem is either a maximal clique, or contained in a maximal clique with the same objective value. We also propose an analytical upper bound for the betweenness centrality of any maximal clique containing a given clique, and employ it to develop a combinatorial branch-and-bound algorithm for solving this problem. Computational performance of the proposed algorithm is then compared with that of a mixed integer programming technique on a test-bed of randomly generated graphs and real-life power-law networks.

Evidence for a Resting State Network Abnormality in Adults Who Stutter.

Neural network-based investigations of stuttering have begun to provide a possible integrative account for the large number of brain-based anomalies associated with stuttering. Here we used resting-state EEG to investigate functional brain networks in adults who stutter (AWS). Participants were 19 AWS and 52 age-, and gender-matched normally fluent speakers. EEGs were recorded and connectivity matrices were generated by LORETA in the theta (4–8 Hz), alpha (8–12 Hz), beta1 (12–20 Hz), and beta2 (20–30 Hz) bands. Small-world propensity (SWP), shortest path, and clustering coefficients were computed for weighted graphs. Minimum spanning tree analysis was also performed and measures were compared by non-parametric permutation test. The results show that small-world topology was evident in the functional networks of all participants. Three graph indices (diameter, clustering coefficient, and shortest path) exhibited significant differences between groups in the theta band and one [maximum betweenness centrality (BC)] measure was significantly different between groups in the beta2 band. AWS show higher BC than control in right temporal and inferior frontal areas and lower BC in the right primary motor cortex. Abnormal functional networks during rest state suggest an anomaly of DMN activity in AWS. Furthermore, functional segregation/integration deficits in the theta network are evident in AWS. These deficits reinforce the hypothesis that there is a neural basis for abnormal executive function in AWS. Increased beta2 BC in the right speech–motor related areas confirms previous evidence that right audio–speech areas are over-activated in AWS. Decreased beta2 BC in the right primary motor cortex is discussed in relation to abnormal neural mechanisms associated with time perception in AWS.

Meditation is associated with increased brain network integration

IntroductionThis study aims to identify novel quantitative EEG measures associated with mindfulness meditation. As there is some evidence that meditation is associated with higher integration of brain networks, we focused on EEG measures of network integration. MethodsSixteen novice meditators and sixteen experienced meditators participated in the study. Novice meditators performed a basic meditation practice that supported effortless awareness, which is an important quality of experience related to mindfulness practices, while their EEG was recorded. Experienced meditators performed a self-selected meditation practice that supported effortless awareness. Network integration was analyzed with maximum betweenness centrality and leaf fraction (which both correlate positively with network integration) as well as with diameter and average eccentricity (which both correlate negatively with network integration), based on a phase-lag index (PLI) and minimum spanning tree (MST) approach. Differences between groups were assessed using repeated-measures ANOVA for the theta (4–8 Hz), alpha (8–13 Hz) and lower beta (13–20 Hz) frequency bands. ResultsMaximum betweenness centrality was significantly higher in experienced meditators than in novices (P = 0.012) in the alpha band. In the same frequency band, leaf fraction showed a trend toward being significantly higher in experienced meditators than in novices (P = 0.056), while diameter and average eccentricity were significantly lower in experienced meditators than in novices (P = 0.016 and P = 0.028 respectively). No significant differences between groups were observed for the theta and beta frequency bands. ConclusionThese results show that alpha band functional network topology is better integrated in experienced meditators than in novice meditators during meditation. This novel finding provides the rationale to investigate the temporal relation between measures of functional connectivity network integration and meditation quality, for example using neurophenomenology experiments.

On the distribution of betweenness centrality in random trees

Betweenness centrality is a quantity that is frequently used to measure how ‘central’ a vertex v is. It is defined as the sum, over pairs of vertices other than v, of the proportions of shortest paths that pass through v. In this paper, we study the distribution of the betweenness centrality in random trees and related, subcritical graph families. Specifically, we prove that the betweenness centrality of the root vertex in a simply generated tree is usually of linear order, but has a mean of order n3/2. We also show that a randomly chosen vertex typically also has linear-order betweenness centrality, and that the maximum betweenness centrality in a simply generated tree is of order n2. We obtain limiting distributions for the betweenness centrality of the root vertex and of a randomly chosen vertex, as well as for the maximum betweenness centrality, and we also show that the centroid has positive probability in the limit to be the vertex of maximum betweenness centrality. Some similar results also hold for subcritical graph classes, which will be briefly discussed. Finally, we study random recursive trees and other families of increasing trees, where the situation is quite different: here, the root betweenness centrality is of quadratic order, as is the maximum betweenness centrality. The betweenness centrality of a random vertex, on the other hand, is again of linear order. Again, we also have limiting distributions upon suitable normalisation.

Optimizing network attacks by artificial bee colony

Over the past few years, the task of conceiving effective attacks to complex networks has arisen as an optimization problem. Attacks are modelled as the process of removing a number k of vertices, from the graph that represents the network, and the goal is to maximise or minimise the value of a predefined metric over the graph. In this work, we present an optimization problem that concerns the selection of nodes to be removed to minimise the maximum betweenness centrality value of the residual graph. This metric evaluates the participation of the nodes in the communications through the shortest paths of the network.To address the problem we propose an artificial bee colony algorithm, which is a swarm intelligence approach inspired in the foraging behaviour of honeybees. In this framework, bees produce new candidate solutions for the problem by exploring the vicinity of previous ones, called food sources. The proposed method exploits useful problem knowledge in this neighbourhood exploration by considering the partial destruction and heuristic reconstruction of selected solutions. The performance of the method, with respect to other models from the literature that can be adapted to face this problem, such as sequential centrality-based attacks, module-based attacks, a genetic algorithm, a simulated annealing approach, and a variable neighbourhood search, is empirically shown.